The Jaiku Opportunity

January 18th, 2009

It was noted on the Google Code Blog that microblogging site Jaiku will be released as Open Source software. The code will target the Google App Engine platform, so that people who want a microblogging site on GAE can build one almost instantly.

Interestingly, the announcement included this sentence:

The new Jaiku Engine will include support for OAuth, and we’re excited about developers using this proven code as a starting point in creating a freely available and federated, open source microblogging platform.

There’s a great opportunity here for Jaiku and its new Open Source clones to join the existing OpenMicroBlogging network. (I suspect that the reference to OAuth, which forms the basis for OMB, might be referring to that.) But it also seems clear that they’ll be looking to Python hackers in the outside community to do the implementation.

There are dangers, too, here. One is that Jaiku creates a custom protocol, effectively cutting off the site and its clones from the OMB network. This would be problematic; it’s unlikely that two competing protocols will get much traction. Another is that the OMB code for the Jaiku engine is implemented in a patch, a fork, or a difficult-to-install plugin, meaning only a minority of Jaiku installations are OMB-connected. I’d love to see the code either part of their core, or in a plugin that’s installed by default (and preferably on by default!).

I’m eager to see engineers from the Jaiku team at Google, and interested Python hackers, get into the OMB discussion on our mailing list. I, for one, look forward to having a connected network of diverse implementations — proprietary and Open Source — that can communicate through simple, clear open standards. Let’s take the steps to make that work.

OMB 0.2 preview

January 9th, 2009

I’ve sent an email to the OMB mailing list with some proposed changes to the OpenMicroBlogging spec. It’s probably too detailed to replicated here in full, but people who are following the spec may want to check it out — and make sure you’re on the list!

A new site and renewed purpose

January 8th, 2009

This is the first post on the new OpenMicroBlogging Web site. OpenMicroBlogging (OMB) is an open standard that allows users on different micro-blogging services to communicate with each other. I think of it like SMTP for email: people on different mail servers can send mail to each other, using the magic of SMTP and universal email addresses. Federated networking systems are much more robust, and have faster levels of innovation, than monolithic systems.

When I launched Identi.ca  in July of 2008, OpenMicroBlogging was implemented from the beginning. It’s one of the things that really captured people’s attention: that we could build an open microblogging network. Because Identi.ca’s software is Open Source (I called it Laconica) , lots of people have been able to set up their own systems and experiment with cross-site subscriptions. There is even one other implementation, OpenMicroBlogger.com, and others on the way.

For good or ill, Identi.ca turned out to be pretty popular, and I’ve spent most of my waking time over the last 6 months on developing Identi.ca-the-site, Laconica-the-software, and Control Yourself, the company that manages them both. As a result, I’ve let the OpenMicroBlogging standard itself slide to somewhere around 98th place in my priority list, right behind eating and dressing myself. I managed to stick the 0.1 version of the spec on a server, but that’s been about the extent of the attention the poor thing has received.

Which is too bad, because there’s lots to do. From the technical side, it’s become clear after millions of messages across the OMB network, that there is a lot of room for improvement in the spec. For example, being able to direct a notice to the attention of a remote user (”@ replies”), sending private messages (”d messages”), and originating a subscription through one’s home server, are all useful features that OMB currently can’t handle. We also need to work out how to synchronize the OMB network and XMPP, using something like the preliminary microblogging XEP.

More importantly, there are community marketing issues that need to be addressed. In order for this effort to succeed, we need to have support from the many Open Source microblogging packages out there — like YonklyTwoorl, Talk.org, WordPress-based Prologue, Six Apart’s freely-distributable-but-not-Open-Source Motion, and others. We also need to do outreach with existing commercial microblogging services like Rejaw, Jaiku, Kwippy, Yammer, and most importantly Plurk and Twitter.

Being part of a larger microblogging network is clearly valuable for users, and thus valuable for the service providers who, uh, provide them with services. And for the software developers who develop software to help the service providers provide services.

All of which is to say: it’s about time to get started. So, I’ve taken some steps to move this OpenMicroBlogging standard ahead.

  1. I’ve started a blog (you’re reading it) to provide a more friendly web face for OpenMicroBlogging than a reStructuredText-converted-to-HTML contextless specification document (more like development notes for Laconica, truth be told). Hopefully this will make it easier to get word out to users, software developers, and service providers on what the advantages of OpenMicroBlogging can be. And we can also use it for news!
  2. I’ve moved the OpenMicroBlogging spec document out of the Laconica source tree and into its own openmicroblogging project on the awesome gitorious version control and project-hosting site. Ideally multiple editors will be managing that document in the future.
  3. I’ve started an OMB mailing list to coordinate software development, implementation, and community outreach. If you are at all interested in OpenMicroBlogging, I heartily suggest th.at you join.

Over the next few weeks, I’m going to be working on a draft for the OpenMicroBlogging 0.2 specification, and I hope to get some input from other interested parties. I expect this will be a quick cycle — stuff that’s hard will be left out, for future implementation. But I’d like to address all critical issues.

Thanks to everyone who’s expressed support or interest in OMB, and my apologies for dragging my feet while life intervened. I hope that all together we can make this a successful protocol, a robust network and an important communications medium.